If you have additional context that is relevant, feed it to the network. Why wou...

slashdave 6 months ago | parent | context | favorite | on: Reasoning in Large Language Models: A Geometric Pe...

If you have additional context that is relevant, feed it to the network. Why wouldn't you? As to the size of the network, this is not a simple benefit, because you need to account for the trade off between model size and training efficiency.